Methods for Proteins Folding

Similar protein sequences have similar 3D structure. Therefore the method of choice for folding a given protein depends on existence of a similar protein whose structure is already known, and on the extent of such similarity.

• When one can find a known-structure protein with good sequence similarity (over $30\%$ amino-acid identity) to the protein we wish to fold, the two proteins will have the same structure. This method is called homology modeling.
• When only less conclusive similarity is available to a known structure, we can use threading as follows. Align our protein to a remotely similar protein whose structure is known. Use the new forced structure as a starting point for finer folding operations.
• When no homology is available one is forced to fold the proteins ab-initio (from scratch). This is hard even in simplistic models. For example we can assume, that the amino acids are divided to two kinds: hydrophobic (water hater) and hydrophilic (water lover). Using this model we can try to build the 3D structure minimizing rejections and maximizing attractions between nearby amino acids. Even solving this model was proven to be NPC, although there are heuristics for it (the best known has $\frac{2}{3}$ approximation ratio).